cortexkit

cortexkit / opencode-magic-context

Public

OpenCode plugin for Magic Context — cache-aware infinite context, cross-session memory, and background history compression for AI coding agents

agent-context agent-framework agent-memory ai ai-agents
17
1
100% credibility
Found Apr 01, 2026 at 17 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

An OpenCode plugin that automatically manages infinite context, compresses conversation history, maintains cross-session memories, and runs overnight maintenance for AI coding agents.

How It Works

1
🔍 Discover better AI memory

You hear about a helper that makes your AI coding buddy remember everything forever, without forgetting old chats.

2
📦 Get the helper

Run a simple command to add it to your coding setup, like installing a useful tool.

3
🧙 Run the setup wizard

Answer a few friendly questions to pick smart thinkers for compressing old talks and remembering key facts.

4
đź’­ Chat with your AI

Your AI now recalls past decisions, project rules, and details across days without you repeating yourself.

5
🌙 Let it dream overnight

While you sleep, it tidies up memories, checks facts against your code, and updates notes automatically.

✨ Endless smart memory

Your AI coding sessions feel magical — infinite history, perfect recall, and always up-to-date project wisdom.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 17 to 17 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is opencode-magic-context?

This TypeScript OpenCode plugin delivers cache-aware infinite context for AI coding agents, handling background history compression, cross-session memory, and structured context injection without interrupting your main agent. Agents get tools like ctx_reduce to drop stale content, ctx_expand to decompress history, ctx_memory for persistent facts, and ctx_search across all layers—plus automatic nudges as context fills. A Tauri dashboard lets you browse memories, inspect sessions, tune configs, and monitor cache hits.

Why is it gaining traction?

Unlike basic compaction in opencode github copilot or agents, it defers drops until caches expire, avoiding wasted prefixes, and runs a lightweight historian plus overnight dreamer for memory consolidation—keeping long sessions viable. The setup wizard auto-detects models, disables conflicting hooks in oh-my-opencode, and supports opencode github integration for enterprise flows. Developers love the zero-config start and commands like /ctx-status for debugging context pressure.

Who should use this?

Backend engineers building complex apps with opencode github agents, where sessions span refactors or architecture shifts. Full-stack teams using opencode plugins list for multi-file edits, tired of context resets. Anyone with opencode github copilot enterprise needing cross-session recall for constraints or decisions.

Verdict

Grab it if you're deep in OpenCode agents—polished docs, wizard, and dashboard punch above the 17 stars and 1.0% credibility score. Early maturity means watch for edge cases in high-volume use, but it transforms agent reliability today.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.